Between Test-Tubes and North Sea: Mesocosms

Abstract

In the marine environment, and the North Sea is no exception, it is often difficult to prove that concentrations of chemicals, elevated above background levels (pollution), cause (reversible of irreversible) effects on the ecosystem. Although the North Sea is one of the most intensively studied seas, our knowledge of its ecology is far from complete (WKP 1986). Measured field data often show large variations, and it is difficult to judge whether extremes in polluted regions fall inside or outside natural ranges. Therefore a widening of our general ecological knowledge is a prerequisite if we are to judge the effects of human interference with the sea. Ecology is a basic science for ecotoxicology. In ecotoxicology the toxic effects of chemical and physical agents on living organisms, especially on populations and communities within ecosystems, are studied together with the transfer pathways of those agents and their interactions with the environment (definition SCOPE, Butler 1978). In this definition the concepts of population and ecosystem take a central position. The key issue is not the protection of individuals of a species, but the conservation of populations and ecosystems. Nevertheless, as in human toxicology, which tries to protect the human individual against undesired effects, most ecotoxicological studies are conducted with individuals of one species. Many authors described the reasons why results of such studies cannot usually be extrapolated to actual field conditions (NAS 1981; Kuiper 1984). Most ecotoxicological tests lack “pollutant and ecological realism” (Blanck et al. 1978). Characteristics of the chemical in the natural environment are not incorporated in the test system. This lack of pollutant realism is related to the chemical form, including speciation and degradation intermediates, concentrations, exposure time to the chemical, and to interactions with other chemicals and physicochemical factors. The influence of these factors on possible effects can be large (e.g., NAS 1981; Kuiper 1984). Lack of ecological realism is often apparent from the choice of test species (sometimes going as far as using tropical freshwater species for predictions in the temperate North Sea!), but in particular from the single-species character of most tests. In the field chemicals act on populations of different species that interact with each other and their surroundings. Interactions in and among species are, for example, apparent in competition and predation. In some ecotoxicological experiments, competition was introduced by dosing chemicals to algal cultures of two or more species (Fisher et al. 1974; Lundy et al. 1984). Heavy metals and PCB’s influence the species composition of these mixed cultures in concentrations that do not effect monocultures of the same species. Changes in species composition are often found as a first effect in pollution research with isolated natural phytoplankton communities (O’Connors et al. 1978). Predation can also be influenced by pollutants. It was shown that exposure to chemicals in concentrations that did not influence monocultures of prey increased their mortality rate, when also exposed to predators, via a change in behaviour caused by the chemical (Ward et al. 1976; Pearson et al. 1981). Via bioaccumulation and biomagnification the predator can be affected. In the field, decreasing populations of top predators are often a first sign of ecosystems malfunctioning. The decrease in numbers of the sandwich tern in the Wadden Sea (Koeman and Van Genderen 1972) and the declining populations of the harbor seal in the same area (Reijnders 1980, 1981, 1983) are well-known examples of this phenomenon.